An investigation of clinical outcome prediction from integrative genomic profiles in ovarian cancer

Lin Zhang, Hui Liu, Tzu Hung Hsiao, Yidong Chen, Yufei Huang

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    6 Scopus citations

    Abstract

    Integrative clinical outcome prediction models that combines gene expression and methylation profiles are investigated in this paper in order to reveal genomic features and models that bear important prognostic value. The models all include the integration and feature selection steps. In the integration step, a method to combine gene expression and methylation profiles is introduced. In the feature selection step, several approaches were investigated including the supervised principal component and the elastic net method to identify genes, whose expression or associate CpG methylation contribute to the clinical outcome. A set of 87 ovarian cancer patients was used in this study to evaluate the proposed methods. The test results showed that the integrative methods improved the prediction performance over those based on gene expression alone.

    Original languageEnglish (US)
    Title of host publicationProceedings 2012 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2012
    Pages103-106
    Number of pages4
    DOIs
    StatePublished - 2012
    Event2012 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2012 - Washington, DC, United States
    Duration: Dec 2 2012Dec 4 2012

    Publication series

    NameProceedings - IEEE International Workshop on Genomic Signal Processing and Statistics
    ISSN (Print)2150-3001
    ISSN (Electronic)2150-301X

    Other

    Other2012 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2012
    Country/TerritoryUnited States
    CityWashington, DC
    Period12/2/1212/4/12

    Keywords

    • clinical outcome prediction
    • data integration
    • elastic net
    • ovarian cancer

    ASJC Scopus subject areas

    • Biochemistry, Genetics and Molecular Biology (miscellaneous)
    • Computational Theory and Mathematics
    • Signal Processing
    • Biomedical Engineering

    Fingerprint

    Dive into the research topics of 'An investigation of clinical outcome prediction from integrative genomic profiles in ovarian cancer'. Together they form a unique fingerprint.

    Cite this